During the process of image generation, internal defects such as pixel defect and pattern noise are created. These defects can be extracted and used to match with reference patterns of digital cameras. Subsequently, the source camera of an image can be identified. Lukas et al. [18–20] have suggested techniques based on pixel non-uniformity (PNU). These methods are a good source to retrieve noise patterns. Nonetheless, these techniques do not work with cropped and different sizes of images [21]. On the other hand, Gharibi et al. [22] proposed a method that involves selecting a region in an image. They reported that each region in an image has different SPN information. The best region is selected based on local information. Then, photo response non-uniformity PRNU of the selected image region is used to analyse the correlation between (PRNU) of the image and PRNU of the camera. Besides that, Xie et al. [23] recommended a technique based on cross correlation between reference PRNU and image residual noise. Through this technique, the sharp peak position of the correlation value can be located and this information can be considered as a unique feature of an image. 

2.2    ImageSOurceIdentificationbasedon CFA Interpolation